Matrix

Added in version 1.6.

class Matrix(*args, **kwargs)
Constructors:

Matrix()

Methods

class Matrix
concat(new_matrix: Matrix) None

Changes the transformation represented by matrix to be the transformation given by first applying transformation given by new_matrix then applying the original transformation.

Added in version 1.6.

Parameters:

new_matrix – a PangoMatrix

free() None

Free a PangoMatrix.

Added in version 1.6.

get_font_scale_factor() float

Returns the scale factor of a matrix on the height of the font.

That is, the scale factor in the direction perpendicular to the vector that the X coordinate is mapped to. If the scale in the X coordinate is needed as well, use get_font_scale_factors.

Added in version 1.12.

get_font_scale_factors() tuple[float, float]

Calculates the scale factor of a matrix on the width and height of the font.

That is, xscale is the scale factor in the direction of the X coordinate, and yscale is the scale factor in the direction perpendicular to the vector that the X coordinate is mapped to.

Note that output numbers will always be non-negative.

Added in version 1.38.

get_slant_ratio() float

Gets the slant ratio of a matrix.

For a simple shear matrix in the form:

1 λ 0 1

this is simply λ.

Added in version 1.50.

rotate(degrees: float) None

Changes the transformation represented by matrix to be the transformation given by first rotating by degrees degrees counter-clockwise then applying the original transformation.

Added in version 1.6.

Parameters:

degrees – degrees to rotate counter-clockwise

scale(scale_x: float, scale_y: float) None

Changes the transformation represented by matrix to be the transformation given by first scaling by sx in the X direction and sy in the Y direction then applying the original transformation.

Added in version 1.6.

Parameters:
  • scale_x – amount to scale by in X direction

  • scale_y – amount to scale by in Y direction

transform_distance() tuple[float, float]

Transforms the distance vector (dx,``dy``) by matrix.

This is similar to transform_point, except that the translation components of the transformation are ignored. The calculation of the returned vector is as follows:

dx2 = dx1 * xx + dy1 * xy;
dy2 = dx1 * yx + dy1 * yy;

Affine transformations are position invariant, so the same vector always transforms to the same vector. If (x1,``y1``) transforms to (x2,``y2``) then (x1``+``dx1,``y1``+``dy1``) will transform to (x1``+``dx2,``y1``+``dy2``) for all values of x1 and x2.

Added in version 1.16.

transform_pixel_rectangle() Rectangle

First transforms the rect using matrix, then calculates the bounding box of the transformed rectangle.

This function is useful for example when you want to draw a rotated PangoLayout to an image buffer, and want to know how large the image should be and how much you should shift the layout when rendering.

For better accuracy, you should use transform_rectangle on original rectangle in Pango units and convert to pixels afterward using extents_to_pixels’s first argument.

Added in version 1.16.

transform_point() tuple[float, float]

Transforms the point (x, y) by matrix.

Added in version 1.16.

transform_rectangle() Rectangle

First transforms rect using matrix, then calculates the bounding box of the transformed rectangle.

This function is useful for example when you want to draw a rotated PangoLayout to an image buffer, and want to know how large the image should be and how much you should shift the layout when rendering.

If you have a rectangle in device units (pixels), use transform_pixel_rectangle.

If you have the rectangle in Pango units and want to convert to transformed pixel bounding box, it is more accurate to transform it first (using this function) and pass the result to extents_to_pixels(), first argument, for an inclusive rounded rectangle. However, there are valid reasons that you may want to convert to pixels first and then transform, for example when the transformed coordinates may overflow in Pango units (large matrix translation for example).

Added in version 1.16.

translate(tx: float, ty: float) None

Changes the transformation represented by matrix to be the transformation given by first translating by (tx, ty) then applying the original transformation.

Added in version 1.6.

Parameters:
  • tx – amount to translate in the X direction

  • ty – amount to translate in the Y direction

Fields

class Matrix
x0

X translation

xx

1st component of the transformation matrix

xy

2nd component of the transformation matrix

y0

Y translation

yx

3rd component of the transformation matrix

yy

4th component of the transformation matrix